Redistribution effects of water tariffs

Nguyen Bich Ngoc, Jacques Teller 24 December 2021

inccatdf <- df[!(is.na(df$inccat)),] %>%
  group_by(inccat) %>%
  summarise(
    count = n(),
    prop = n() / nrow(df),
    income_avr = mean(income),
    income_min = min(income),
    income_max = max(income),
    inceqa_avr = mean(inceqa),
    inceqa_min = min(inceqa),
    inceqa_max = max(inceqa)
  )

inccatdf

## # A tibble: 4 x 9
##   inccat     count   prop income_avr income_min income_max inceqa_avr
##   <fct>      <int>  <dbl>      <dbl>      <int>      <int>      <dbl>
## 1 precarious   164 0.0949      1144.        125       2250      8928.
## 2 modest       781 0.452       1894.       1250       3250     16022.
## 3 average      566 0.328       3216.       2250       4750     23687.
## 4 higher       216 0.125       4759.       3750       5250     30933.
## # ... with 2 more variables: inceqa_min <dbl>, inceqa_max <dbl>

## Warning: Removed 192 row(s) containing missing values (geom_path).

## Warning: Removed 150 row(s) containing missing values (geom_path).

# Correlation between water consumption and household income should use spearman?????

cor.test(df$csmptv, df$income, method = "pearson")

## 
##  Pearson's product-moment correlation
## 
## data:  df$csmptv and df$income
## t = 15.729, df = 1726, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
##  0.3121384 0.3946480
## sample estimates:
##      cor 
## 0.354082

cor.test(df$csmptv, df$income, method = "spearman")

## Warning in cor.test.default(df$csmptv, df$income, method = "spearman"):
## Cannot compute exact p-value with ties

## 
##  Spearman's rank correlation rho
## 
## data:  df$csmptv and df$income
## S = 536353649, p-value < 2.2e-16
## alternative hypothesis: true rho is not equal to 0
## sample estimates:
##       rho 
## 0.3763062

# Correlation between water consumption and income per equivalent adult should use spearman?????

cor.test(df$csmptv, df$inceqa, method = "pearson")

## 
##  Pearson's product-moment correlation
## 
## data:  df$csmptv and df$inceqa
## t = 1.4269, df = 1726, p-value = 0.1538
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
##  -0.01285165  0.08134826
## sample estimates:
##        cor 
## 0.03432454

cor.test(df$csmptv, df$inceqa, method = "spearman")

## Warning in cor.test.default(df$csmptv, df$inceqa, method = "spearman"):
## Cannot compute exact p-value with ties

## 
##  Spearman's rank correlation rho
## 
## data:  df$csmptv and df$inceqa
## S = 803896839, p-value = 0.006707
## alternative hypothesis: true rho is not equal to 0
## sample estimates:
##        rho 
## 0.06519613

summary(df$avrprc)

##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
##   4.200   4.595   4.690   5.000   4.850  12.549

summary(df$avrprc[df$poorest == 1])

##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
## 

summary(df$avrprc[df$inccat == "precarious"])

##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max.    NA's 
##   4.222   4.605   4.734   5.152   4.915  10.592       1

summary(df$subs[df$inccat == "precarious"])

##     Min.  1st Qu.   Median     Mean  3rd Qu.     Max.     NA's 
## -68.1035  -4.7756   0.5658  -4.0094   8.3744  27.4910        1

summary(df$mgnprc)

##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
##   1.158   4.458   4.458   3.982   4.458   4.658

summary(df$mgnprc[df$poorest == 1])

##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
## 

summary(df$mgnprc[df$inccat == "precarious"])

##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max.    NA's 
##   1.158   4.153   4.458   3.754   4.458   4.658       1

## 3.5. changing fixed  -----

### new cvd ------